initial release

README.md

@@ -1,3 +1,39 @@
----
-license: mit
----
+---
+language:
+- "de"
+tags:
+- "german"
+- "token-classification"
+- "pos"
+- "dependency-parsing"
+datasets:
+- "universal_dependencies"
+license: "mit"
+pipeline_tag: "token-classification"
+---
+
+# bert-base-german-upos
+
+## Model Description
+
+This is a RoBERTa model pre-trained with [UD_German-HDT](https://github.com/UniversalDependencies/UD_German-HDT) for POS-tagging and dependency-parsing, derived from [bert-base-german-cased](https://huggingface.co/bert-base-german-cased). Every word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech).
+
+## How to Use
+
+```py
+from transformers import AutoTokenizer,AutoModelForTokenClassification
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/bert-base-german-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/bert-base-german-upos")
+```
+
+or
+
+```py
+import esupar
+nlp=esupar.load("KoichiYasuoka/bert-base-german-upos")
+```
+
+## See Also
+
+[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa models
+

config.json

@@ -0,0 +1,355 @@
+{
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "ADJ",
+    "1": "ADJ+NOUN",
+    "2": "ADP",
+    "3": "ADP+DET",
+    "4": "ADV",
+    "5": "AUX",
+    "6": "B-ADJ",
+    "7": "B-ADP",
+    "8": "B-ADP+DET",
+    "9": "B-ADV",
+    "10": "B-AUX",
+    "11": "B-CCONJ",
+    "12": "B-DET",
+    "13": "B-DET+NOUN",
+    "14": "B-INTJ",
+    "15": "B-NOUN",
+    "16": "B-NOUN+PROPN",
+    "17": "B-NUM",
+    "18": "B-PRON",
+    "19": "B-PROPN",
+    "20": "B-PROPN+PROPN",
+    "21": "B-PUNCT",
+    "22": "B-SCONJ",
+    "23": "B-VERB",
+    "24": "B-X",
+    "25": "CCONJ",
+    "26": "CCONJ+NOUN",
+    "27": "DET",
+    "28": "DET+NOUN",
+    "29": "I-ADJ",
+    "30": "I-ADP",
+    "31": "I-ADP+DET",
+    "32": "I-ADV",
+    "33": "I-AUX",
+    "34": "I-CCONJ",
+    "35": "I-DET",
+    "36": "I-DET+NOUN",
+    "37": "I-INTJ",
+    "38": "I-NOUN",
+    "39": "I-NOUN+PROPN",
+    "40": "I-NUM",
+    "41": "I-PRON",
+    "42": "I-PROPN",
+    "43": "I-PROPN+PROPN",
+    "44": "I-PUNCT",
+    "45": "I-SCONJ",
+    "46": "I-VERB",
+    "47": "I-X",
+    "48": "INTJ",
+    "49": "NOUN",
+    "50": "NOUN+PROPN",
+    "51": "NOUN+X",
+    "52": "NUM",
+    "53": "PART",
+    "54": "PRON",
+    "55": "PROPN",
+    "56": "PROPN+PROPN",
+    "57": "PROPN+X",
+    "58": "PUNCT",
+    "59": "SCONJ",
+    "60": "SYM",
+    "61": "VERB",
+    "62": "X"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "ADJ": 0,
+    "ADJ+NOUN": 1,
+    "ADP": 2,
+    "ADP+DET": 3,
+    "ADV": 4,
+    "AUX": 5,
+    "B-ADJ": 6,
+    "B-ADP": 7,
+    "B-ADP+DET": 8,
+    "B-ADV": 9,
+    "B-AUX": 10,
+    "B-CCONJ": 11,
+    "B-DET": 12,
+    "B-DET+NOUN": 13,
+    "B-INTJ": 14,
+    "B-NOUN": 15,
+    "B-NOUN+PROPN": 16,
+    "B-NUM": 17,
+    "B-PRON": 18,
+    "B-PROPN": 19,
+    "B-PROPN+PROPN": 20,
+    "B-PUNCT": 21,
+    "B-SCONJ": 22,
+    "B-VERB": 23,
+    "B-X": 24,
+    "CCONJ": 25,
+    "CCONJ+NOUN": 26,
+    "DET": 27,
+    "DET+NOUN": 28,
+    "I-ADJ": 29,
+    "I-ADP": 30,
+    "I-ADP+DET": 31,
+    "I-ADV": 32,
+    "I-AUX": 33,
+    "I-CCONJ": 34,
+    "I-DET": 35,
+    "I-DET+NOUN": 36,
+    "I-INTJ": 37,
+    "I-NOUN": 38,
+    "I-NOUN+PROPN": 39,
+    "I-NUM": 40,
+    "I-PRON": 41,
+    "I-PROPN": 42,
+    "I-PROPN+PROPN": 43,
+    "I-PUNCT": 44,
+    "I-SCONJ": 45,
+    "I-VERB": 46,
+    "I-X": 47,
+    "INTJ": 48,
+    "NOUN": 49,
+    "NOUN+PROPN": 50,
+    "NOUN+X": 51,
+    "NUM": 52,
+    "PART": 53,
+    "PRON": 54,
+    "PROPN": 55,
+    "PROPN+PROPN": 56,
+    "PROPN+X": 57,
+    "PUNCT": 58,
+    "SCONJ": 59,
+    "SYM": 60,
+    "VERB": 61,
+    "X": 62
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "task_specific_params": {
+    "upos_multiword": {
+      "ADJ+NOUN": {
+        "selbst\u00e4ndigeDenken\u00b8": [
+          "selbst\u00e4ndige",
+          "Denken\u00b8"
+        ],
+        "\u00f6ffentlichenInternetcaf\u00e9s": [
+          "\u00f6ffentlichen",
+          "Internetcaf\u00e9s"
+        ]
+      },
+      "ADP+DET": {
+        "A.": [
+          "AN",
+          "DEM"
+        ],
+        "Am": [
+          "An",
+          "dem"
+        ],
+        "Ans": [
+          "An",
+          "das"
+        ],
+        "Beim": [
+          "Bei",
+          "dem"
+        ],
+        "F\u00fcrs": [
+          "F\u00fcr",
+          "das"
+        ],
+        "Im": [
+          "In",
+          "dem"
+        ],
+        "Ins": [
+          "In",
+          "das"
+        ],
+        "Ums": [
+          "Um",
+          "das"
+        ],
+        "Unterm": [
+          "Unter",
+          "dem"
+        ],
+        "Vom": [
+          "Von",
+          "dem"
+        ],
+        "Zum": [
+          "Zu",
+          "dem"
+        ],
+        "Zur": [
+          "Zu",
+          "der"
+        ],
+        "am": [
+          "an",
+          "dem"
+        ],
+        "ans": [
+          "an",
+          "das"
+        ],
+        "aufs": [
+          "auf",
+          "das"
+        ],
+        "beim": [
+          "bei",
+          "dem"
+        ],
+        "durchs": [
+          "durch",
+          "das"
+        ],
+        "f\u00fcrs": [
+          "f\u00fcr",
+          "das"
+        ],
+        "hinterm": [
+          "hinter",
+          "dem"
+        ],
+        "hinters": [
+          "hinter",
+          "das"
+        ],
+        "im": [
+          "in",
+          "dem"
+        ],
+        "ins": [
+          "in",
+          "das"
+        ],
+        "ums": [
+          "um",
+          "das"
+        ],
+        "unterm": [
+          "unter",
+          "dem"
+        ],
+        "unters": [
+          "unter",
+          "das"
+        ],
+        "vom": [
+          "von",
+          "dem"
+        ],
+        "vorm": [
+          "vor",
+          "dem"
+        ],
+        "vors": [
+          "vor",
+          "das"
+        ],
+        "z.": [
+          "zu",
+          "dem"
+        ],
+        "zum": [
+          "zu",
+          "dem"
+        ],
+        "zur": [
+          "zu",
+          "der"
+        ],
+        "\u00dcbers": [
+          "\u00dcber",
+          "das"
+        ],
+        "\u00fcbers": [
+          "\u00fcber",
+          "das"
+        ]
+      },
+      "CCONJ+NOUN": {
+        "sowieInternetcaf\u00e9s": [
+          "sowie",
+          "Internetcaf\u00e9s"
+        ]
+      },
+      "DET+NOUN": {
+        "dieCr\u00e8me": [
+          "die",
+          "Cr\u00e8me"
+        ],
+        "dieFort\u00e9-Familie": [
+          "die",
+          "Fort\u00e9-Familie"
+        ]
+      },
+      "NOUN+PROPN": {
+        "Highend-ModellCli\u00e9": [
+          "Highend-Modell",
+          "Cli\u00e9"
+        ],
+        "McKinsey-BeraterRen\u00e9": [
+          "McKinsey-Berater",
+          "Ren\u00e9"
+        ],
+        "NachfolgemodellCli\u00e9": [
+          "Nachfolgemodell",
+          "Cli\u00e9"
+        ]
+      },
+      "PROPN+PROPN": {
+        "AlexanderArtop\u00e9": [
+          "Alexander",
+          "Artop\u00e9"
+        ],
+        "ClausS\u00f8rensen": [
+          "Claus",
+          "S\u00f8rensen"
+        ],
+        "Jean-LouisGass\u00e9e": [
+          "Jean-Louis",
+          "Gass\u00e9e"
+        ],
+        "JoelleRichardi\u00e9re": [
+          "Joelle",
+          "Richardi\u00e9re"
+        ],
+        "LeD\u00e9aut": [
+          "Le",
+          "D\u00e9aut"
+        ]
+      }
+    }
+  },
+  "tokenizer_class": "BertTokenizerFast",
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22c187760763df4a0ba59d50560fd2464f373b3e771ffd63b4bf2a81430eae04
+size 434244299

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

supar.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6767137a559c0cda3f40dacaa05190e965713588b2e3d5aea8a355f830e9cad
+size 513838054

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "tokenizer_class": "BertTokenizerFast"}